Inquiry about how to connect an existing, locally hosted Weaviate instance with pre-loaded embeddings as a knowledge base in the Agno framework without re-ingesting data

Hi,

For the agent’s knowledge, we need to use a locally hosted Weaviate vector database that already contains pre-generated embeddings.

We explored the knowledge base options you provided, but we couldn’t find a way to connect our existing Weaviate instance as a knowledge base without ingesting new data through your interface.

Since our Weaviate instance is already populated with data and embeddings, we want to connect to it as-is, without re-uploading or re-indexing anything.

Can you please guide us on how to achieve this?

Hey @sham_sundar, thanks for reaching out and supporting Agno. I’ve shared this with the team, we’re working through all requests one by one and will get back to you soon.
If it’s urgent, please let us know. We appreciate your patience!

Hi @Monali,

We are facing the same scenario, albiet a different vector database(milvus), but we would also like to know conceptually how to connect an existing vector database and want to use as-is.

Hi @sham_sundar and @Latshawk

Thanks for raising your issues. We are planning on improving our Knowledge functionality and use cases like these are great.

If I understood your question correctly, you want to attach an existing Weaviate instance to an Agno KnowledgeBase.

Would you mind taking a look at this cookbook please. It shows how to connect a Weaviate DB to a KnowledgeBase. You won’t need to call the load() function if all of the documents you need are already embedded.
Please let me know if this satisfies your requirements and if not, what the issues are that you run into.